Machine-Learned Control Customization for Dynamic Resource Prioritization

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing operating systems of computing devices are designed for general-purpose use and offer limited opportunities for users to customize their behavior, leading to inefficient resource allocation and performance for specific use types.

Innovation Solution

A customization component dynamically reconfigures the control system using a machine-trained model to prioritize resource allocation based on use types, adapting to changing use scenarios through supervised or reinforcement learning, and incorporating user feedback and historical patterns.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If the control system is designed for general-purpose use, then it can serve a broad class of users, but it cannot efficiently allocate resources for specific use types

Engineering Contradiction:

Improveadaptability to different use typesVSAvoidresource allocation efficiency

Core Design Contradiction:

Adaptability or versatilityVSProductivity

Solution Approach 1:

The control system dynamically adapts its resource allocation policies based on detected use types. Instead of being static, the system continuously monitors usage patterns and adjusts prioritization rules in real-time, allowing it to optimize resource allocation for the current use type while maintaining general-purpose functionality across different scenarios

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system changes operational parameters (resource allocation weights, prioritization rules) based on the detected use type. By adjusting these parameters dynamically, the control system can optimize performance for specific use types such as gaming, productivity, or media consumption while still supporting multiple different use scenarios

Inventive Principle:

Principle #35Parameter changes

2Ease of operation

If the control system offers limited customization opportunities, then it maintains simplicity, but it cannot optimize performance for specific user needs

Engineering Contradiction:

Improvecustomization capabilityVSAvoidcomputing performance

Core Design Contradiction:

Ease of operationVSProductivity

Solution Approach 1:

The control system automatically detects use types and adjusts resource allocation without requiring manual user configuration. The system serves itself by monitoring its own operational state and making intelligent adjustments based on detected patterns, eliminating the need for complex user-facing customization interfaces while still achieving optimization

Inventive Principle:

Principle #25Self-service

3Stability of the object's composition

If the control system statically allocates resources, then it maintains stability, but it cannot adapt to changing use scenarios during a computing session

Engineering Contradiction:

Improvesystem stabilityVSAvoiddynamic adaptation to use type changes

Core Design Contradiction:

Stability of the object's compositionVSAdaptability or versatility

Solution Approach 1:

The control system transitions from static to dynamic resource allocation by continuously monitoring usage patterns and adjusting priorities in real-time. This dynamic approach allows the system to adapt to changing use scenarios during a computing session while maintaining operational stability through controlled, incremental adjustments

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system implements feedback loops that continuously monitor resource usage patterns and detect use type changes. Based on this feedback, the control system automatically adjusts resource allocation policies, creating a closed-loop system that adapts to changing conditions while maintaining overall system stability

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS12498678B2Customizing a control system for different use types using a machine-trained model for generating and the modifying

Publication Date: 2025.12.16 MICROSOFT TECHNOLOGY LICENSING LLC

AI summary

A technique is described for customizing a control system (e.g., an operating system) of a computing system for different use types. The different use types are associated with different ways of using the computing system. The technique improves performance of computing functions prioritized at a particular time. For instance, the technique improves performance by reducing the number of interruptions experienced by the computing system when executing prioritized applications. Further, in some implementations, the technique dynamically adapts to changing uses over the course of a single computing session. The technique performs this task by dynamically reassessing the use type that most accurately characterizes the current use of the computing system. The technique operates using a discrete computer program, machine-trained functionality, or some combination thereof. Some implementations of the machine-trained functionality use reinforcement learning.